Hyperhomocysteinemia is associated with cognitive impairment in multiple sclerosis

Hyperhomocysteinemia (HHcy) has been associated with cognitive impairment in various neurological diseases. Cognitive impairment occurs early in multiple sclerosis (MS). Conflicting data have been reported regarding plasma total homocysteine (tHcy)
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  J Neurol (2008) 255:64–69DOI 10.1007/s00415-007-0668-7 ORIGINAL COMMUNICATION C.RussoF.MorabitoF.LuiseA.PiromalliL.BattagliaA.VinciV.Trapani MarcoP.MorabitoF.CondinoA.QuattroneUmberto Aguglia Hyperhomocysteinemia is associated withcognitive impairment in multiple sclerosis    J   O   N    2   6   6   8 Introduction Cognitive impairment occurs early in multiple sclerosis(MS) and affects 30–70% ofpatients [1,2].Differences in the nature and extent ofcognitive dysfunction in MSare related to a variety offactors including clinicalcourse,duration ofillness,disability,treatment,and le-sion loads demonstrated by brain magnetic resonanceimaging (MRI) [1–4]. Homocysteine (Hcy) is a sulfur-containing metabo-lite ofmethionine that can become elevated in the plasma Received:12 February 2007Received in revised form:8 May 2007Accepted:5 June 2007Published online:20 December 2007F.Luise · A.Piromalli · V.Trapani LombardoThrombosis and Hemostasis’s CentreC.Russo · L.Battaglia · A.Vinci · P.MorabitoNeurologic UnitAzienda Ospedaliera Bianchi-Melacrino-MorelliReggio Calabria,Italy F.MorabitoHematology Unit Azienda OspedalieraCosenza,Italy A.Quattrone · U.AgugliaChair ofNeurology Magna Graecia University Catanzaro,Italy Marco · F.Condino · A.QuattroneInstitute ofNeurological SciencesNational Research CouncilMangone (Cosenza),Italy U.Aguglia MD (  )Magna Graecia University ofCatanzaroRegional Epilepsy Center at Azienda OspedalieraVia Melacrino89100 Reggio Cal.,Italy Tel.:0965/397971Fax:0965/ ■ Abstract Hyperhomocystein-emia (HHcy) has been associatedwith cognitive impairment in vari-ous neurological diseases.Cogni-tive impairment occurs early inmultiple sclerosis (MS).Conflictingdata have been reported regardingplasma total homocysteine (tHcy)levels in MS patients,and the im-pact ofHHcy on cognitive impair-ment in MS is not known.Thisstudy investigated whether plasmatotal homocysteine levels are in-creased in MS and ifHHcy is asso-ciated with cognitive impairmentin MS.We compared tHcy levels in94 patients with MS and 53 healthy age-matched controls.We used aneuropsychological test battery that included the Raven’s ColouredProgressive Matrices,the VisualSearch Test,the Trail Making Test Aand B,the Immediate and DelayedRecall ofa Short Story,the 30Paired Word Associates,the Rey-Osterrieth Complex Figure Test,and the Semantic and Verbal Flu-ency Tests.Clinical (sex,age,typeofMS,relapse,disease duration,co-existing disease,smoking habit,and physical disability) and labora-tory variables (HHcy,low serumlevels offolate and vit.B12,MTHFRgenotype) were evaluated for theirability to predict cognitive impair-ment.The mean tHcy was higher inpatients (13.19µmol/L,SD5.58)than in controls (9.81µmol/L,SD2.53;p<0.001).Univariateanalysis determined the followingfactors to be associated with cogni-tive impairment:higher age at ob-servation,chronic progressivecourse ofdisease,longer diseaseduration,moderate or severe physi-cal disability,and frequency of HHcy.With multivariate regressionanalysis,there remained a signifi-cant association only between fre-quency ofHHcy and cognitive im-pairment ( β 0.262,p=0.01).Weconclude that tHcy levels are in-creased in MS and that HHcy is as-sociated with cognitive impairmentin this disease. ■ Key words cognitiveperformance · homocysteine ·multiple sclerosis ·neuropsychology   65 as a result ofgenetic or nutrient-related disturbances [5,6].Recent studies investigated the link between hyperho-mocysteinemia (HHcy) and the risk ofdementia [7,8]and cognitive impairment in older adults [9,10]. Studies comparing plasma total homocysteine(tHcy) concentrations between MS patients and con-trols have yielded conflicting findings [11–13].The rela- tionship between HHcy and cognitive impairment inMS is unknown.The aim ofthis study was twofold.First,we sought to test the hypothesis that patients with MShave higher tHcy levels than healthy controls.Second,we sought to determine whether HHcy is associatedwith cognitive impairment in MS. Methods ■ Patients and controls One hundred thirty-seven consecutive patients referred to our Neu-rology Unit in the last 2 years were considered for enrollment.Inclu-sion criteria were diagnosis ofMS according to the revised McDon-ald’s criteria [14] and no corticosteroid treatment for at least 3months.Patients in relapse were admitted,and corticosteroid treat-ment was initiated soon after completion ofthe study.Exclusion cri-teria were vitamin B12 or folate supplementation within the preced-ing year,presence ofpsychiatric illness,including severe depressionand treatment with neuroleptic,antidepressant,or antiepilepticdrugs.The presence or absence ofdepression was determined usingthe Beck Depression Inventory [15].Ninety-four of137 patients (52 women,42 men;age 20–69 years,median 36 years) matched the inclusion or exclusion criteria and en-tered the study.The clinical types ofMS were:relapsing-remitting(n=77 patients,81.9%) and chronic progressive (n=17,18.1%;n=15 secondary progressive MS,n=2 primary progressive MS).Allpatients underwent cerebral and cervical MRI examination aftertriple-dose gadolinium-diethylenetriaminepentaacetic acid withinthe 7 days preceding or following tHcy determination.At the time of the study,17 of94 patients (18.1%) had MRI evidence ofrelapse.Themean educational level was 9.18±4 years (range=5–18 years,me-dian=8 years).The duration ofillness (difference between age at ex-amination and age at the first clinical manifestation) ranged between1.2 months and 41.12 years (mean 10.2±8.9 years).Twenty-five of94patients (26.6%) were habitual smokers.Coexisting disease was pres-ent in 7 of94 patients (7.4%) and included essential hypertension(n=4),deep vein thrombosis (n=1),ischemic cardiomyopathy (n=1),and diabetes mellitus (n=3).Physical disability was rated us-ing Kurtzke’s Expanded Disability Status Scale (EDSS) score [16],andthe patients were separated into two groups according to an EDSS cut-offvalue of3.5.Sixty-five patients (69.1%) had an EDSS score ≤3.5,and 29 (30.9%) had a score >3.5.Healthy volunteers (blood donors) were recruited from theThrombosis and Hemostasis Centre (Azienda Ospedaliera,ReggioCal.).None ofthe controls were treated with vitamin B12 or folatesupplementation.All subjects (patients and controls) gave informedconsent before inclusion in the study. ■ Neuropsychological assessment The neuropsychological evaluation was performed in the week pre-ceding or following tHcy determination.The evaluation required ap-proximately 2 hours and was performed in two sessions.To minimizethe effect offatigue,periods ofrest were provided during testing.Theevaluation was performed using the Raven’s Coloured ProgressiveMatrices [17],the Visual Search Test [18],the Trail Making Test A andB [19],the Immediate and Delayed Recall ofa Short Story [18],the 30Paired Word Associates [20],the Rey-Osterrieth Complex Figure Test[21],the Semantic Verbal Fluency Test [18],and the Phonemic VerbalFluency Test [22].Results were compared with Italian publishednorms [17–22].Individual test performance was considered abnor- mal when it was two or more SDs below the control mean.To analyzedifferences in cognitive dysfunction,we divided patients into threesubgroups based on the number oftests failed:unimpaired (0–2 failedtests),mildly impaired (3–5 failed tests),and moderately impaired(>5 failed tests). ■ Laboratory assessments Blood samples were assessed for tHcy,serum vitamin B12,and serumfolate.All blood samples were obtained from fasting subjects from 8:00to 9:00 a.m.Blood was collected from patients and controls by the samepersonnel,in the same setting,at the Thrombosis and Hemostasis’sCentre.Plasma samples were stored at –80 °C until analysis.Plasma wasobtained by centrifuging whole blood at 3000 × g for 20 min at 4 °Cwithin 2 hours after collection.tHcy in plasma was measured with a mi-croplate enzyme immunoassay (BioRad Laboratories,Inc,Hercules,CA,USA) [23].HHcy was defined as tHcy ≥15µmol/L.Serum folate andvitamin B12 levels were assayed in 24 patients and 35 controls using acommercially available kit (Vitros Immunodiagnostic System;Ortho-Clinical Diagnostics,Milan,Italy).The C677T mutation ofthe methyl-ene tetrahydrofolate reductase (MTHFR) gene was examined in 57 pa-tients by PCR-RFLP ofDNA samples [24] using the enzyme HinfI. ■ Statistical analysis Statistical comparisons involving binary variables were performed us-ing 2-way tables for the Fisher exact test and multi-way tables for thePearson chi-square test.The nonparametric Kruskal-Wallis andMann-Whitney U test were used to analyze differences between two ormore groups.The Spearman rank correlation coefficient (rho) wasused to assess the strength ofthe straight-line association between thevariables.Since the dependent variable was restricted to more than twocategories,linear regression,which estimates the coefficients ofthelinear equation involving one or more independent variables that bestpredict the value ofthe dependent variable,was used for multivariateanalysis.The strength ofthe association between each independentvariable and the dependent variable was expressed as the standardizedregression coefficient beta.The variables with probability values <0.10were eligible for inclusion as predictor variables in the multivariate re-gression analysis.We constructed a series ofmultiple regression mod-els ofincreasing complexity (models #1–4,Table4) in which we in-cluded all ofthe variables that were significantly associated withcognitive impairment at the univariate analysis.A p<0.05 was consid-ered significant for all statistical calculations.All statistical calcula-tions were performed using the statistical package SPSS for Windows,release 11.5,2002 (SPSS UK,Working,Surrey,United Kingdom). Results Participant demographics are shown in Table1.The twogroups had comparable age,sex,and smoking habits. ■ tHcy levels The mean plasma tHcy level was significantly higheramong MS patients (13.19µmol/L) than controls  66 (9.81µmol/L,p<0.001,Table1).The frequency ofpatho-logical values was also significantly higher in patientsversus controls (MS:33/94,35.10%;controls:2/53,3.8%;p<0.0001).There were no significant differences inmean levels or pathological values ofserum vitamin B12or serum folate (Table1). ■ Variables associated with cognitive impairment Cognitive impairment was present in 34/94 patients(36.17%).It was mild in 24/94 (25.53%) and moderatein 10/94 (10.63%).The MTHFR genotype was C/C in 22,C/T in 25,and T/T in 10 patients.Levels ofplasma tHcy significantly correlated with the number offailed cogni-tive tests (Fig.1A) and were significantly higher amongcases with a moderately impaired cognitive function(Fig.1B).Univariate analysis (Table2) determined thatthe following variables were significantly associatedwith cognitive impairment:higher age at observation,chronic progressive course,longer duration ofdisease,moderate or severe disability (EDSS >3.5),higherplasma tHcy levels,and frequency ofHHcy.Sex,relapse,coexisting disease,smoking habit,educational level,fre-quency oflow serum folate,and MTHFR genotype didnot significantly correlate with cognitive impairment.Multivariate regression analysis with the variables age atobservation,chronic progressive course,duration ofdis-ease,EDSS >3.5,and frequency ofHHcy,determinedonly frequency ofHHcy to be significantly associatedwith cognitive dysfunction (Table4;beta 0.262,p=0.01). ■ Cognitive impairment in patients with HHcy Compared to MS patients with normal tHcy levels,pa-tients with HHcy had abnormalities in nonverbal rea-soning (Raven’s Progressive Matrices),visual attention(visual search test,Trail Making Test A and B),visual-spatial memory (recall on the Rey-Osterrieth Complex PatientsHealthy controlspNumber of subjects9453Sex (F/M)52/4228/25NSAge (years), mean (SD)36.63 (10.36)37.15 (12.06)NSSmoking habit25 (26.6%)13 (24.5%)NSPlasma tHcy (µmol/L), mean (SD)13.19 (5.58)9.81 (2.53)<0.001Frequency of HHcy33 (35.10%)2 (3.8%)<0.0001Serum folate* (nmol/L), mean±SD (range)14.3±12.7 (3.0–52)13.8±8.2 (3.8–42)NSFrequency of low serum folate*63NSSerum vitamin B12* (pmol/L), mean±SD 273±112 (98–578)291±109 (119–532)NS(range)Frequency of low serum vitamin B12*00NS* Determination performed in 24 patients and 35 controls F  female; HHcy  hyperhomocysteine; M male; SD standard deviation; NS  nonsignificant; tHcy  total homocysteine Table1 Participant demographics and results of laboratory tests Fig.1 Relationship between levels of total plasma homocysteine and cognitiveperformance. A Spearman correlation coefficient (rho). Significant positive correla-tion between levels of total plasma homocysteine (tHcy) and failed cognitive tests. B Mann-Whitney U test. Significantly higher levels of tHcy in patients with moder-ately impaired cognitive function     P    l   a   s   m   a    t    H   c   y      µ    m   o    l    /    L    (   m   e   a   n   ±   s   e   m    ) 18161412unimpairedmildly impaired moderately impairedp=0.0050     P    l   a   s   m   a    t    H   c   y    (      µ    m   o    l    /    L    ) number of failed cognitive tests010203040501234RHO 0.346p=0.0015678910 ab cognitive function  67 Figure Test),and visual-spatial ability (copy ability onRey-Osterrieth Complex Figure Test).In contrast,verbalfluency (phonemic and semantic verbal fluency) andverbal memory (short story and 30 paired word associ-ates) were not significantly compromised in the HHcy group (Table3). Discussion In agreement with previous reports [12–13],we found that patients with MS had increased plasma tHcy levels.This is in contrast to another study [11] that found nosignificant differences with regard to serum Hcy be-tween MS patients and controls.However,this differencein findings may be explained by the smaller sample sizein the former study (38 patients and 20 healthy blooddonors).Homocysteine is produced during methioninemetabolism,and is removed either by conversion to cys-teine (with vitamin B6 as a cofactor) or by remethylationto methionine (with vitamin B12 and folate as cofac-tors).Deficiency ofvitamin B6,B12,and folate may cause HHcy.We found that patients and controls hadequivalent levels ofvitamin B12 and folate.Although wedid not determine levels ofvitamin B6,other authors[13] demonstrated no significant differences in theplasma levels ofvitamin B6 between MS patients andcontrols.Collectively,these data suggest that HHcy may be caused by increased production rather than de-creased removal ofHcy.In previous studies,HHcy was shown to be a risk fac-tor for dementia and Alzheimer’s disease [3] and was as-sociated with worse cognitive performance and cogni-tive decline in nondemented elderly people [5,6].We showed that HHcy is significantly associated with cog-nitive impairment and that there was a significant posi-tive correlation between plasma tHcy levels and cogni-tive impairment in MS patients.Increased tHcy levelshave been associated with depression,reduced memory function,and reduced constructional ability in nonde-mented elderly subjects [6,25].A previous pilot study  [26] examined the correlation between tHcy levels and VariablesUnimpairedMildly impairedModerately impairedp(n=60)(n=24)(n=10)Male sex, n (%)24 (40)11 (45.8)7 (77.8)NSAge in years, mean±SD34.7±8.537.2±12.247.0±10.40.018Educational level; mean±SD9.0±4.08.9±3.99.0±4.3NSChronic progressive course, n (%)8 (13.3)2 (8.3)7 (70.0)<0.0001Relapse, n (%)10 (16.7)5 (20.8)2 (20.0)NSDisease duration in years, mean±SD8.0±7.011.9±10.318.9±10.60.002Coexisting disease, n (%)3 (5.0)3 (12.5)1 (10.0)NSSmoking, n (%)15 (25.0)7 (29.2)3 (30.0)NSEDSS >3.5, n (%)13 (21.7)9 (37.5)7 (70.0)0.007Plasma tHcy (µmol/L), mean±SD12.4±5.813.5±4.817.1±4.50.005HHcy, n (%)14 (23.3)11 (45.8)8 (80.0)0.001Low folate*, n (%)4 (26.7)2 (25.0)0 (0)NSMTHFR** genotype C/T or T/T, n (%)21 (63.6)12 (66.7)2 (33.3)NS* Determination performed in 24 patients; ** Determination performed in 57 patients EDSS  Kurtzke’s Expanded Disability Status Scale; HHcy  hyperhomocysteinemic status; MTHFR methylene tetrahy-drofolate reductase; NS  not significant; SD standard deviation; tHcy  plasma total homocysteine Table2 Univariate analysis of risk factors of cogni-tive dysfunction in 91 MS patients Table3 Relationship between homocysteine status and cognitive performanceCognitive testsPatientsNormal tHcyHyper tHcypRaven’s Coloured Progressive Matricesnormal/abnormal59/224/90.001Visual search testnormal/abnormal59/225/80.003Trail Making Test Anormal/abnormal42/1914/190.016Trail Making Test Bnormal/abnormal41/2014/190.028Short storynormal/abnormal43/1822/110.81630 paired word associatesnormal/abnormal56/522/110.003Rey-Osterrieth Complex Figure Test (recall)normal/abnormal31/309/240.031Rey-Osterrieth Complex Figure Test (copy)normal/abnormal55/624/90.039Semantic verbal fluencynormal/abnormal59/229/40.179Phonemic verbal fluencynormal/abnormal54/726/70.234 tHcy  total plasmatic homocysteine; NS  not significant  68 clinical functioning in MS patients.Results showed apositive correlation between tHcy levels and the EDSSand a negative correlation with the MS Functional Com-posite test (MSFC),indicating worse functioning in pa-tients with HHcy.In addition,there was a negative cor-relation between tHcy levels and cognition subtests of the MSFC,suggesting a relationship between HHcy andcognitive impairment.These results should be inter-preted with some caution since these authors did notcorrect for educational level,which has recently beendemonstrated to affect the cognitive performance ofMSpatients [27].We found no significant differences in theeducational level between patients whose cognition wasunimpaired,mildly impaired,and severely impaired.In MS patients,primary language functions,immedi-ate and implicit memory,and verbal intellectual skillsseem to be preserved,whereas information processingabilities,complex visuospatial tasks,conceptual reason-ing,sustained attention,and working memory are oftenimpaired [28].In our study,the cognitive battery in-cluded tests ofattention,but processing speed andworking memory assessment were not fully covered,soworking memory was not adequately explored.Never-theless,we found that the most severely affected func-tions in patients with HHcy were nonverbal reasoning,visual attention,visual-spatial memory and visual-spa-tial ability.Verbal fluency and verbal memory were notsignificantly different between patients with HHcy andnormal tHcy.These cognitive dysfunctions seem to bedifferent from those observed in subcortical dementiaand nondemented older adults and may be explained onthe basis ofa “multiple disconnection syndrome”[28].In a previous longitudinal study [7] in which tHcy levels were not evaluated,the extent ofcognitive impair-ment in MS was associated with age,disease course,anddisability.In our study,the association between HHcy and cognitive impairment appeared to be independentofage,progressive disease course,and degree ofphysi-cal disability,suggesting a direct relationship betweenHHcy and cognitive dysfunction.We also consideredfunctional mutations ofthe MTHFR gene,which havebeen associated with increased levels oftHcy,as a possi-ble independent factor determining cognitive impair-ment.However,our findings do not support a role forfunctional mutation ofthe MTHFR gene and CI in MS.Nonetheless,it must be emphasized that the cross-sec-tional nature ofour analysis does not allow us to infer acausality link between HHcy and cognitive impairmentin MS,and future prospective cohort studies to test sucha causal relationship are warranted.Studies investigating possible predictors ofcognitiveimpairment in MS have focused on MRI measurementsas indicators ofneuropathological involvement;suchstudies have shown consistent correlations between cog-nitive impairment and brain lesion load or brain atro-phy [29].HHcy is associated with increased risk ofvas-cular disease.It could provoke “silent”subcorticallacunar infarcts,increase the white-matter lesion load inMS patients,and provoke additional cognitive impair-ment.On the other hand,it has been hypothesized thatHcy produces cognitive impairment in the elderly via adirect neurotoxic effect [6],and a significant associationhas been demonstrated between HHcy and cortical at-rophy and dementia [30].We did not investigate the cor-relations between MRI data and cognitive impairmentand tHcy levels.Thus,we are unable to confirm or refutethese hypotheses,and further studies are needed to clar-ify this issue.As an alternative hypothesis,HHcy shouldbe considered as an epi-phenomenon secondary to thedegenerative process in MS as well as in dementia andnormal ageing.In conclusion,HHcy is independently associatedwith cognitive impairment in MS.Although physicianstreat cases with HHcy,further prospective and interven-tion studies are needed to determine whether loweringthe plasma Hcy level can reduce the risk ofcognitive de-cline and dementia in MS. Table4 Multiple regression models associated with cognitive functionDependent variable: cognitive functionCovariatesUnadjustedModel 1Model 2Model 3Model 4HHcy( β =0.379), P<0.0001( β =0.311), P=0.002( β =0.290), P=0.004( β =0.264), P=0.008( β =2.262), P=0.010Age...( β =0.255), P=0.010( β =0.09), P=0.5( β =0.07), P=0.6( β =0.06), P=0.6Duration of disease......( β =0.233), P=0.086( β =0.197), P=0.15( β =0.198), P=0.15Chronic progressive course.........( β =0.141), P=0.18( β =0.128), P=0.3EDSS............( β =0.02), P=0.8Data are reported as standardized regression coefficients ( β ) and P value. EDSS  Kurtzke’s Expanded Disability Status Scale; HHcy  hyperhomocysteinemic status
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks